1 2005-12-27 Matthias Clasen <mclasen@redhat.com>
3 * glib/gdatasetprivate.h: Add a cast to silence compiler
4 warnings. (#321978, Andrew Paprocki)
6 Partial fix for bug #32937, Bogdan Nicula.
8 * configure.in: Check for malloc.h
9 * glib/gslice.c: Don't include malloc.h unconditionally.
11 2005-12-27 Manish Singh <yosh@gimp.org>
13 * tests/run-collate-tests.sh: export LC_COLLATE so it takes.
14 (#324950, Dan Yefimov)
16 2005-12-26 Matthias Clasen <mclasen@redhat.com>
18 * glib/gslice.c: On Win32, include process.h (#325015, Kazuki
21 2005-12-25 Matthias Clasen <mclasen@redhat.com>
23 * AUTHORS: Update my email
25 * tests/slice-test.c: Fix C99isms. (#324950, Dan Yefimov)
27 2005-12-21 Matthias Clasen <mclasen@redhat.com>
29 * README.in: Add some notes on when 'make check' may fail.
31 2005-12-20 Matthias Clasen <mclasen@redhat.com>
35 * glib/gthreadpool.c (g_thread_pool_set_sort_function): New function
36 to sort tasks pushed into a threadpool. (#324479, Martyn Russell)
38 * tests/threadpool-test.c: Test this.
40 Tue Dec 20 18:14:14 2005 Tim Janik <timj@imendio.com>
42 * glib/gslice.[hc]: added mem_error() and mem_assert() to test and
43 handle errors without depending on gmessage.c which might not be
44 setup when the error occours.
45 removed G_SLICE_CONFIG_ALWAYS_FREE config option, fixed the code so
46 always freeing can be achieved by adjusting the working set time to
47 0 with G_SLICE_CONFIG_WORKING_SET_MSECS.
48 added G_SLICE_CONFIG_COLOR_INCREMENT to test different color increments
49 (mainly 0 and 1). reduced the minimum block size to 128 bytes, to
50 minimize wastage if small amounts of differently sized structrues are
51 allocated, this does come at a performance cost of roughly 5% though.
52 fixed up block alignment calculation, so it works for varying
53 block sizes. only use strerror() not g_strerror() since the latter
54 depends on working GQuark and GSlice.
55 mem_error(): implemented in terms of fprintf and vfprintf.
57 * tests/slice-color.c: new program to test cache colorization effects.
59 * tests/slice-test.c: trade G_SLICE_CONFIG_ALWAYS_FREE for 0 duration
60 G_SLICE_CONFIG_WORKING_SET_MSECS.
62 2005-12-17 Matthias Clasen <mclasen@redhat.com>
64 * glib/goption.c (parse_short_option): Set an error in all
65 failure cases. (#324332, Tim-Philipp Müller)
67 2005-12-17 Sebastian Wilhelmi <seppi@seppi.de>
69 * glib/gatomic.c: Fix memory barrier position in g_atomic_int_get
70 and g_atomic_pointer_get. Add g_atomic_int_set and
71 g_atomic_pointer_set implementations for the !DEFINE_WITH_MUTEXES &&
72 G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case, as well as defining them
73 as functions (additionally to the macros in the header) for the
74 !G_ATOMIC_OP_MEMORY_BARRIER_NEEDED case.
76 2005-12-16 Matthias Clasen <mclasen@redhat.com>
78 * glib/gmem.c (g_allocator_new): Don't return a pointer to
79 a const struct, since apps expect to be able to modify it.
80 (#324179, J. Ali Harlow)
82 Tue Dec 13 10:13:32 2005 Tim Janik <timj@imendio.com>
84 * glib/gatomic.h: added g_atomic_pointer_set() and g_atomic_int_set()
85 for systems where the initialization of atomic variables requires a
88 2005-12-09 Matthias Clasen <mclasen@redhat.com>
90 * configure.in: Bump version
92 * === Released 2.9.1 ===
94 2005-12-09 Alexander Larsson <alexl@redhat.com>
98 Use G_GNUC_WARN_UNUSED_RESULT on list functions that return
101 2005-12-08 Matthias Clasen <mclasen@redhat.com>
105 2005-12-07 Martyn Russell <martyn@imendio.com>
107 * glib/gasyncqueue.c:
108 - Call g_queue_insert_sorted() instead of duplicating the code.
109 - Call g_queue_sort() instead of duplicating the code.
110 - Invert sort function results to make sure the same sort function
111 gives the same results across glist, gslist, gqueue and
114 * tests/asyncqueue-test.c:
115 - Updated the sort function to reflect the example in the
116 documentation for gasyncqueue.c.
118 2005-12-07 Martyn Russell <martyn@imendio.com>
120 * docs/reference/glib/glib-sections.txt:
121 * docs/reference/glib/tmpl/linked_lists_double.sgml:
122 * docs/reference/glib/tmpl/linked_lists_single.sgml:
125 - Added g_list_insert_sorted_with_data () and
126 g_slist_insert_sorted_with_data ().
127 - Removed the extra check in g_list_sort() and g_slist_sort() for
128 GCompareDataFunc vs. GCompareFunc.
130 2005-12-07 Tor Lillqvist <tml@novell.com>
132 * glib-gettextize.in: Look up prefix at run-time on Win32,
133 assuming the standard directory structure with glib-gettextize in
136 2005-12-06 Matthias Clasen <mclasen@redhat.com>
138 * glib/gmem.h: Exempt GTK+ from the mem chunk deprecation, since
139 we need GTK+ 2.8 to compile against GLib 2.10.
141 2005-12-05 Manish Singh <yosh@gimp.org>
143 * tests/run-collate-test.sh: set LC_COLLATE, not LANG, to be sure
144 to override any user settings.
146 2005-12-05 Matthias Clasen <mclasen@redhat.com>
148 * configure.in: Revert an accidential commit
150 2005-12-05 Matthias Clasen <mclasen@redhat.com>
152 * glib/gutf8.c: Documentation updates. (#323291, Morten Welinder)
154 * tests/Makefile.am (TESTS_ENVIRONMENT):
155 * tests/gobject/Makefile.am (TESTS_ENVIRONMENT): Set
156 MALLOC_CHECK_ and MALLOC_PERTURB_
158 * tests/run-collate-tests.sh: Run the collation tests explicitly
159 in en_US locale. (#320463)
161 * glib/gmem.h: Really deprecate GMemChunk.
168 * glib/gstrfuncs.c: Add versioned deprecation docs.
170 2005-12-05 Martyn Russell <martyn@imendio.com>
172 * docs/reference/glib/glib-sections.txt:
173 * glib/gasyncqueue.[ch]:
174 - Added support for sorting async queues by with _push_sorted(),
175 _push_sorted_unlocked(), _sort() and _sort_unlocked() (#323047).
178 * tests/asyncqueue-test.c:
179 - Added test case for gasyncqueue.c
181 Mon Dec 5 15:53:20 2005 Tim Janik <timj@imendio.com>
183 * glib/gslice.c: implement chain walking for arbitrary ->next pointer
184 offsets in g_slice_free_chain_with_offset() based on a patch by behdad
185 in bug 323178. moved time consuming logic from g_slice_free() out of
186 the inner loop, so g_slice_free_chain_with_offset() provides a real
187 performance benefit over g_slice_free1() now.
189 * glib/gslice.h: renamed g_slice_free_chain() to
190 g_slice_free_chain_with_offset(). implemented g_slice_free_chain() as
191 a type-safe macro as suggested in bug 323178.
192 simplified the macro implementation of g_slice_free() and implemented
193 it in a type safe manner for all compliers as suggested by Morten
194 Welinder <mortenw@gnome.org>.
199 * glib/glib.symbols: s/g_slice_free_chain/g_slice_free_chain_with_offset/
201 2005-12-05 Matthias Clasen <mclasen@redhat.com>
203 * glib/gasyncqueue.c: Add some docs.
205 * tests/libmoduletestplugin_a.c: Fix compiler warnings.
207 * glib/gatomic.c: In the ia64 implementation, use
208 __sync builtin without _si or _di suffix. (#321229,
209 Stanislav Brabec, patch by Andreas Schwab)
211 2005-12-04 Behdad Esfahbod <behdad@gnome.org>
213 * glib/gslice.h: Remove comma at the end of enum.
215 2005-12-04 Matthias Clasen <mclasen@redhat.com>
217 Handle multiple user names with the same UID better.
218 (#319535, Laszlo Peter)
220 * glib/gutils.c (g_get_any_init_do): When determining user
221 data, first look up $LOGNAME. If the UID doesn't match
222 getuid(), fall back to the current behaviour of looking
223 up the user data based on getuid().
225 2005-12-04 Matthias Clasen <mclasen@redhat.com>
227 * glib/gmacros.h (G_GNUC_WARN_UNUSED_RESULT): Add a macro
228 to make gcc warn if a function result is ignored. (#145466,
229 Arjan van de Ven, Alex Larsson)
231 * glib/gmem.h: Add the new attribute to g_realloc and
234 2005-12-03 Matthias Clasen <mclasen@redhat.com>
237 * glib/gcache.h: Deprecate g_cache_value_foreach. (#322956,
240 * glib/gmappedfile.c: Make mapping of empty files
243 * glib/gfileutils.c: Don't fork a new process just to
244 fix the permissions of the created temp file. (#321318,
245 Alexis S. L. Carvalho)
247 2005-12-02 Matthias Clasen <mclasen@redhat.com>
249 * README.in: Add a note about Solaris threads.
252 * configure.in: Remove support for Solaris threads.
253 (#136971, Sebastian Wilhelmi, patch by Andrew Paprocki)
255 2005-12-02 Matthias Clasen <mclasen@redhat.com>
257 * configure.in: Define G_GUINT64_CONSTANT in analogy to
258 G_GINT64_CONSTANT. (#322568, Andrew Paprocki)
260 2005-12-02 Matthias Clasen <mclasen@redhat.com>
262 * glib/gslice.c: Win32 portability fixes and C99ism removal,
263 pointed out by Kazuki Iwamoto. (#323052)
265 2005-12-02 Matthias Clasen <mclasen@redhat.com>
267 * glib/gatomic.c (g_atomic_int_add): Add a missing volatile for
268 the IA64 implementation.
270 Fri Dec 2 16:18:09 2005 Tim Janik <timj@imendio.com>
272 * glib/gslice.c: some naming and type size fixups.
274 Fri Dec 2 13:08:58 2005 Tim Janik <timj@gtk.org>
276 * glib/gslice.h (g_slice_free): reworked GCC-specific type-safe macro
277 variant into something less verbose, but digestible for gcc-3.4.
279 Fri Dec 2 10:55:07 2005 Tim Janik <timj@imendio.com>
281 * tests/slice-test.c: extended to perform the benchmarking on the old
282 memchunk code if 'O' is selected.
284 * tests/memchunks.c: new file which contains the old GLib mem chunks
285 implementation with prefix old_mem_chunk_.
287 * tests/Makefile.am: added memchunks.c
289 Fri Dec 2 00:16:59 2005 Tim Janik <timj@imendio.com>
291 * glib/gslice.c: improved gettimeofday() timer resolution by maintaining
292 acceptable load distribution of the syscall.
293 (allocator_get_magazine_threshold): reduce minimum/single-thread magazine
294 sizes to approximately page_size/2.5 to avoid excessive startup allocations.
296 2005-12-01 Matthias Clasen <mclasen@redhat.com>
298 * tests/gobject/ifacecheck.c (test_iface_base_init): Don't
299 double-free base interface members.
301 * tests/Makefile.am: Remove duplicate variable.
303 Thu Dec 1 17:32:46 2005 Tim Janik <timj@imendio.com>
305 * glib/gslice.[hc]: new slice allocator implementation.
307 * tests/slice-test.c: added random slice allocation test.
309 * glib/gthread.[hc]: removed newly added private thread mem API.
311 * glib/gthreadinit.h:
314 * glib/gmem.c: divided glib threading initialisation into three phases,
315 initialisation where private keys and messaging are not available (only
316 needed by gmem.c), initialisation without messaging but private keys
317 available (gslice.c, gmessage.c), and full fledged initialisers that
318 server the rest of glib. initialisation functions got renamed to reflect
319 the limitations of their corresponding phases.
321 * glib/gmem.c: removed memchunk code, defer allocations to
324 * glib/gmem.[hc]: removed g_slice_* skeletons.
326 * glib/glib.symbols: added g_slice_* symbols.
328 * configure.in: check for availability of posix_memalign(3), memalign(3)
331 * glib/Makefile.am: added gslice.[hc].
333 2005-12-01 Tor Lillqvist <tml@novell.com>
335 * glib/gstdio.c (g_stat): In the Win32 implementation, strip
336 trailing slash(es) for non-root folders. stat() fails if non-root
337 folders are specified with trailing slashes. It's too much hassle
338 to demand that callers strip such slashes themselves, especially
339 as it is easy to get it wrong and strip the slash of a root
341 (g_rename): On NT-based Windows, use MoveFileEx() with
342 MOVEFILE_REPLACE_EXISTING to better match Unix behaviour.
344 2005-11-28 Matthias Clasen <mclasen@redhat.com>
346 Fix G_STMT_START / G_STMT_END on Solaris. (#321972,
349 * configure.in: Check whether do { } while (0) works.
351 * glib/gmacros.h: Use do { } while (0) for G_STMT_START /
352 G_STMT_END if it works.
354 2005-11-28 Matthias Clasen <mclasen@redhat.com>
356 * glib/gthread.c (g_static_rw_lock_wait, g_static_rw_lock_signal):
357 * glib/gnode.c (g_node_depth_traverse_level):
358 * glib/gmem.c (g_allocator_new):
359 * glib/ghash.c (g_hash_table_unref, g_hash_table_destroy)
360 (g_hash_table_foreach_remove_or_steal): Silence compiler
363 2005-11-27 Matthias Clasen <mclasen@redhat.com>
365 * glib/gunicollate.c (g_utf8_collate_key_for_filename):
366 Don't read beyond len. (#322520, Christian Persch)
368 Wed Nov 23 17:34:01 2005 Tim Janik <timj@imendio.com>
370 * glib/gdataset.c: access datalist flags via atomic pointer access
371 functions, instead of acquiring the dataset lock. this is faster and
372 also matches the atomic pointer readouts in gdatalistprivate.h and
373 direct pointer modifications required by gobject.c.
375 Wed Nov 23 13:35:31 2005 Tim Janik <timj@gtk.org>
377 * glib/gdataset.c: streamlined initialization code somewhat,
378 removed GData node cache.
380 2005-11-22 Matthias Clasen <mclasen@redhat.com>
382 * glib/ghash.c (g_hash_table_ref):
383 (g_hash_table_unref): Mark these as new API.
385 Avoid double locking in g_intern_string (#322133,
388 * glib/gdataset.c (g_quark_from_string_internal): New
389 internal function which factors out common parts of
390 g_quark_from[_static]_string.
391 (g_quark_from_string, g_quark_from_static_string):
392 Use g_quark_from_string_internal.
393 (g_intern_string, g_intern_static_string): Use
394 g_quark_from_string_internal, and only take the
396 (g_quark_new): Don't store the strings shifted by -1
397 in the g_quarks array.
398 (g_quark_to_string): Adapt to the previous change.
400 Tue Nov 22 14:04:26 2005 Tim Janik <timj@imendio.com>
404 g_hash_table_new_full(): create hash tables with a ref count of 1.
405 g_hash_table_ref(): atomically ref_count+=1
406 g_hash_table_unref(): atomically ref_count-=1, destroys hash table
407 when refcount reaches 0.
408 g_hash_table_destroy(): just destroy keys and values, unref by 1.
409 g_hash_table_insert():
410 g_hash_table_replace(): assert ref_count>0.
413 * glib/gatomic.c: added 'volatile' qualifier to all atomic pointer and
416 2005-11-20 Behdad Esfahbod <behdad@gnome.org>
418 * glib/guniprop.c (g_unichar_get_mirror_char): Remove unused
419 variables. (#321984, Andrew Paprocki)
421 2005-11-18 Matthias Clasen <mclasen@redhat.com>
423 * configure.in: Bump version
425 2005-11-17 Matthias Clasen <mclasen@redhat.com>
427 * === Released 2.9.0 ===
429 * glib/glib.symbols: Updates
431 * glib/gutf8.c (g_utf8_offset_to_pointer): Add some performance
432 hints to the docs. (Paolo Borelli)
436 * glib/gthread.c (g_thread_foreach): Mark as new api.
438 * README.in: Updates.
442 * glib/gdate.c (g_date_set_time_t):
443 (g_date_set_time_val): Add functions to set a date from
444 a time_t and from a GTimeVal.
445 (g_date_set_time): Deprecate. (#314067, Roger Leigh)
448 * tests/utf8-pointer.c: Unit tests for g_utf8_pointer_to_offset
449 and g_utf8_offset_to_pointer.
451 * glib/gutf8.c (g_utf8_pointer_to_offset)
452 (g_utf8_offset_to_pointer): Handle negative offsets, and use
453 "stutter stepping" for going backwards. (#320638, Larry
458 * glib/gthread.c: const correctness fixes, found
459 by Arjan van de Ven and gcc.
461 2005-11-16 Behdad Esfahbod <behdad@gnome.org>
463 * docs/reference/glib/tmpl/unicode.sgml: Correct typo on mentioning
464 @G_UNICODE_BREAK_UNKNOWN that should be #G_UNICODE_BREAK_UNKNOWN.
466 2005-11-09 Behdad Esfahbod <behdad@gnome.org>
468 * glib/guniprop.c: Use bit hacks instead when checking a general
469 category value against multiple values.
471 * glib/gutf8.c: Change ISO10646 to Unicode in docs.
473 2005-11-10 Simos Xenitellis <simos@gnome.org>
475 * configure.in: Added tt (Tatar) to ALL_LINGUAS.
477 2005-11-05 Tor Lillqvist <tml@novell.com>
479 * glib/giowin32.c (g_io_win32_set_flags): Don't call g_warning().
480 (#320688) Instead set error and return failure.
482 2005-11-07 Matthias Clasen <mclasen@redhat.com>
484 * glib/gmain.c (g_main_dispatch): Don't call cb_funcs->unref
485 while holding the context lock. (#320886, Andy Wingo)
487 2005-11-07 Billy Biggs <vektor@dumbterm.net>
489 * tests/refcount/Makefile.am: Add a missing reference to libglib
490 in the LDADD for this test.
492 2005-11-07 Matthias Clasen <mclasen@redhat.com>
494 * glib/gmem.c (g_allocator_new): We need to set n_preallocs to a
495 nonzero value, otherwise GTK+ 2.8 breaks when compiled against
496 GLib 2.9. (#320755, Luca Ferretti)
498 2005-11-04 Matthias Clasen <mclasen@redhat.com>
500 * glib/Makefile.am: Apply a patch from Behdad Esfahbod to
501 use a faster lookup table for g_unichar_get_mirror_char().
502 * glib/gmirroringtable.h: The new table.
503 * glib/gunichartables.h: Remove bidi_mirroring_table.
504 * glib/gen-unicode-tables.pl: Don't generate the mirroring
506 * glib/glib-mirroring-tab/*: Sources for the program
507 which generated gmirroringtable.h.
509 * glib/glist.c: Avoid some code duplication.
511 * glib/gscanner.h: Include gdataset.h. (#320322)
513 * glib/gdebug.h: add new GDebugFlag for fatal_criticals
514 * glib/gmessages.c: (_g_debug_init): handle G_DEBUG=fatal_criticals,
515 to help find critical warnings in applications. (#320017,
518 2005-11-02 Tor Lillqvist <tml@novell.com>
520 * glib/glib.symbols: Remove large amount of trailing whitespace
521 from one line. Remove #ifdef G_UNDEFINED from around
522 g_hash_table_new and g_hash_table_new_full symbols, what was the
523 purpose of that? Add the g_slice functions.
525 Tue Nov 1 16:24:20 2005 Tim Janik <timj@imendio.com>
527 * glib/gmem.[hc]: prepared deprecation of GMemChunk and GAllocator.
528 added g_slice_*() API to allocate and cache small bits of memory.
529 an actuall allocator implementation for g_slice_*() is still pending.
531 * glib/gthread.[hc]: changes from a patch by Matthias Clasen.
532 changed GRealThread list to use in-structure *next; fields instead
533 of GSList, in order for thread iteration to not depenend on g_slice_*()
535 _g_thread_mem_private_get():
536 _g_thread_mem_private_set(): added accessors for private memory,
537 needed because the ordinary GPrivate implementation relies on GArray
538 and GSList and therefore indirectly on working g_slice_*() allocations.
541 g_thread_foreach(): new public API function to loop over all existing threads.
555 removed GAllocator and free list usages and accompanying locks.
556 use g_slice_*() API to allocate and cache small bits of memory.
558 * glib/ghook.h: removed GMemChunk field from public API.
561 * glib/glist.h: deprecate allocator API, provide _free1() for consistency.
563 * glib/gnode.h: deprecate allocator API.
565 * glib/gmain.c: reordered GPollRec fields so g_slice_free_chain() can
566 be used for poll rec lists.
568 * glib/grel.c: removed mem chunk usage, and allocated tuples via g_slice_*().
569 g_relation_destroy(): free all tuples from the all_tuples hash table,
570 this effectively maintains the life time track keeping of tuples.
571 g_relation_delete_tuple(): free tuples which are removed from the
572 all_tuples hash table. this fixes a temporary leak that was present
573 in the memchunk code until the destruction of the relation.
575 2005-10-29 Matthias Clasen <mclasen@redhat.com>
577 * tests/convert-test.c: Add some tests for conversions between
578 UTF-8, UCS-4 and UTF-16.
580 * glib/gutf8.c (g_utf8_to_ucs4, g_utf8_to_utf16): Fix handling
581 of len == -1, noticed by Morten Welinder.
583 2005-10-27 Erdal Ronahi <erdal.ronahi@gmail.com>
585 * configure.in: Added ku (Kurdish) to ALL_LINGUAS
587 2005-10-26 Matthias Clasen <mclasen@redhat.com>
589 * glib/gutf8.c (g_ucs4_to_utf8): Don't set items_read twice
590 in the error case, and add some documentation. (#319806, Morten
593 2005-10-19 Manish Singh <yosh@gimp.org>
595 * configure.in: Use AC_CHECK_FUNCS for _NSGetEnviron, to get the
596 config.h symbol automatically. Fixes bug #313731.
598 2005-10-19 Tor Lillqvist <tml@novell.com>
600 * glib/gwin32.c (g_win32_get_package_installation_directory):
601 Return a g_strdup()ed copy of the value stored in the hash table,
602 so that it can be g_free()d without leaving a dangling pointer in
603 the hash table. (#319232)
605 2005-10-06 Matthias Clasen <mclasen@redhat.com>
607 * glib/gunicollate.c (g_utf8_collate_key_for_filename): Handle
608 all-zero sequences correctly. (#317930, Sebastien Bacher)
610 2005-10-05 Matthias Clasen <mclasen@redhat.com>
612 * glib/gmarkup.c (g_markup_escape_text): Doc addition
614 * Makefile.am: Add Collation tests.
616 * tests/collate/*: Inputs and expected outputs for collation tests.
618 * tests/run-collate-tests.sh: Script to run collation tests.
620 * tests/unicode-collate.c (main): Rework slightly to make
621 it usable in unit tests. Also test g_utf8_collate_key_for_filename().
623 2005-10-01 Behdad Esfahbod <behdad@gnome.org>
625 * docs/reference/glib/tmpl/unicode.sgml:
626 * glib/gen-unicode-tables.pl:
628 * glib/gunichartables.h:
630 * tests/casefold.txt:
631 * tests/casemap.txt: Updated to Unicode 4.1. There are five new
632 GUnicodeBreakType types. That may break some applications, like
635 2005-09-26 Matthias Clasen <mclasen@redhat.com>
637 * glib/gstrfuncs.c (g_ascii_strtoull): Add details to the
638 docs. (#314393, Matthew F. Barnes)
641 * glib/gprintf.h: Remove g_snprintf() and g_vsnprintf(), since
642 they are already declared in glib.h. This doesn't break documented
643 use of gprintf.h, but should probably be pointed out in the
644 release notes for 2.10. (#314232, Behdad Esfahbod)
646 Tue Sep 20 13:16:04 2005 Tim Janik <timj@imendio.com>
648 * glib/gpattern.c (g_pattern_ph_match): applied significant recursion
649 complexity optimization, based on a patch from Matthias Clasen.
651 * tests/patterntest.c: more tests, mostly from matthias.
653 2005-09-20 Matthias Clasen <mclasen@redhat.com>
655 * glib/gqueue.c (g_queue_insert_sorted): Correct the docs.
656 (#316703, Mark Drago)
658 Mon Sep 19 17:23:23 2005 Tim Janik <timj@imendio.com>
660 * glib/gpattern.c: applied a patch from matthias which checks on the
661 upper bound of GPatternSpec length to optimize matches.
664 * tests/patterntest.c: added more match cases.
666 2005-09-16 Tor Lillqvist <tml@novell.com>
668 * glib/gstrfuncs.c (g_ascii_strcasecmp, g_ascii_strncasecmp): Add
669 warning to doc comment that these functions should not be used on
670 encodings like CP932.
672 2005-09-14 Matthias Clasen <mclasen@redhat.com>
674 * tests/keyfile-test.c: Add a test for grup names of length 1.
676 * glib/gkeyfile.c (g_key_file_line_is_group): Accept group names
677 of length 1. (#316309)
679 2005-09-12 Matthias Clasen <mclasen@redhat.com>
681 * glib/gmarkup.c (g_markup_escape_text): Clarify docs.
683 2005-09-11 Sebastian Wilhelmi <seppi@seppi.de>
685 * tests/refcount/Makefile.am (INCLUDES): Link the the refcount
686 tests to the system thread library $(G_THREAD_LIBS). Fixes #313744
689 2005-09-11 Kjartan Maraas <kmaraas@gnome.org>
691 * glib/gmain.c: (g_child_watch_prepare), (g_child_watch_check),
692 (child_watch_helper_thread): Remove some dead code. Closes
695 2005-09-07 Tor Lillqvist <tml@novell.com>
697 * glib/Makefile.am: Create also a console version of the
698 gspawn-win32-helper program, gspawn-win32-helper-console.exe.
699 It's otherwise identical to gspawn-win32-helper.exe, except marked
700 as a console application (linked without the -mwindows option).
702 * glib/gspawn-win32.c (do_spawn_directly, do_spawn_with_pipes):
703 Drop the dont_wait parameter. Its truth value correlated 100% with
704 the NULLness of the exit_status parameter anyway, so it's enough
705 to check whether exit_status is NULL. Invert the sense of the
706 dont_return_handle parameter and rename it to do_return_handle, to
707 make the code easier to read by avoiding double negations.
709 (g_spawn_sync_utf8, g_spawn_async_with_pipes_utf8): Modify calls
710 to do_spawn_with_pipes() accordingly.
712 (do_spawn_with_pipes): If we have a console, use the console
713 version of the helper program, otherwise use the GUI one. This
714 avoids extra console windows opening up in some situations. (In
715 case a console application uses the GUI gspawn-win32-helper.exe to
716 spawn another console application we would get a separate console
717 for the spawned console application).
719 * glib-zip.in: Distribute also gspawn-win32-helper-console.exe.
721 2005-09-05 Matthias Clasen <mclasen@redhat.com>
723 * glib/gmappedfile.c (g_mapped_file_new): Report an error
724 if the file is too large. (#315275, Kjartan Maraas)
726 * glib/gkeyfile.c (g_key_file_load_from_fd): The return value
727 of read() is signed. (#315273, Kjartan Maraas)
729 2005-08-31 Tor Lillqvist <tml@novell.com>
731 * glib/gutils.h: Wrapping atexit() is a bad idea on Windows, where
732 the EXE and each DLL have their own atexit function chains.
734 #define g_atexit as atexit instead. This means it has a
735 better chance of doing what the caller wants. For instance,
736 gtkhtml calls g_atexit() registering a function in gtkhtml
737 itself. This caused a crash when g_atexit() was implemented as a
738 function in the GLib DLL. The gtkhtml DLL was already unloaded by
739 the time the GLib DLL got unloaded.
741 * glib/gutils.c: #undef the #define mentioned above, to also get a
742 real g_atexit() into the DLL for backward compatibility. Document
743 the Windows behaviour of g_atexit(), and document the varying ways
744 atexit() can behave in the context of dynamically loaded modules
747 2005-08-31 Matthias Clasen <mclasen@redhat.com>
751 * glib/gdataset.c: Add string interning functions.
753 2005-08-28 Matthias Clasen <mclasen@redhat.com>
755 * glib/giochannel.c: Unify some near-duplicate strings. (#314654,
758 2005-08-26 Matthias Clasen <mclasen@redhat.com>
760 * configure.in: Bump version to 2.9.0
762 2005-08-25 Tor Lillqvist <tml@novell.com>
764 Make also the g_spawn*() functions take parameters in the GLib
765 file name encoding, i.e. UTF-8, on Windows. Has no impact on Unix
766 API or ABI. Like the other GLib API that was earlier changed to
767 use UTF-8 on Windows, the names of the functions that take UTF-8
768 have _utf8 suffixes added by using preprocessor macros in the
769 header file. The old names are kept for functions with the old
770 behaviour, taking parameters in the system codepage, for DLL ABI
773 * glib/gspawn.h: On Win32 add the suffix _utf8 to the names of the
774 g_spawn*() functions.
776 * glib/gspawn-win32.c: Use wide-char API on NT-based
777 Windows. Convert parameters from UTF-8 to wide chars (NT) or
778 system codepage (Win9x) and call the C library _wspawn*() or
779 spawn*() functions respectvely. Add DLL ABI stability versions
780 that take parameters in the system codepage.
782 * glib/gspawn-win32-helper.c: On NT-based Windows use the
783 wide-char versions of argv and envp, and use wide-char API to
784 change directory and spawn the program to run. Remove the verbose
785 debugging output, it was too complex to modify for the wide-char
786 features. (Just add temporary debugging printouts if needed, no
787 need to have them permanently in the source.)
789 * glib/gspawn.c: Corresponding documentation updates.
791 * glib/glib.symbols: Corresponding changes: Mark the ABI stability
792 symbols as PRIVATE, add the new _utf8-suffixed ones.
794 2005-08-24 Stepan Kasal <kasal@ucw.cz>
796 * glib/gtypes.h (G_MININT64): Cast the constant to gint64; it is
797 guint64 otherwise and that can produce warnings about comparison
798 between signed and unsigned.
800 2005-08-23 Matthias Clasen <mclasen@redhat.com>
802 * glib/gutils.c: Fix the crt_externs.h include.
804 2005-08-23 Stepan Kasal <kasal@ucw.cz>
806 * NEWS: Fix spelling of my first name.
808 2005-08-23 Matthias Clasen <mclasen@redhat.com>
812 * === Released 2.8.1 ===
816 2005-08-20 Hans Breuer <hans@breuer.org>
818 * glib/makefile.msc.in : link with ws2_32.lib
820 2005-08-18 Tor Lillqvist <tml@novell.com>
822 * configure.in: Check for <sys/wait.h>
824 * glib/gbacktrace.c: Include <sys/wait.h> on if HAVE_SYS_WAIT_H.
826 2005-08-18 Ross Burton <ross@burtonini.com>
829 Optimise single-character insertions.
834 * tests/string-test.c:
835 Add tests for new optimisation, and fix a leak.
837 2005-08-17 Matthias Clasen <mclasen@redhat.com>
839 * configure.in: Check for crt_externs.h and _NSGetEnviron.
841 * glib/gutils.c: On Darwin, include crt-externs.h and
842 define environ using _NSGetEnviron(). (#313731)
844 2005-08-16 Stepan Kasal <kasal@ucw.cz>
846 * glib/gutils.c (g_get_any_init): Move the body of the big if...
847 (g_get_any_init_do): ... to this new function.
848 (g_get_any_init): Declare as inline.
849 (g_get_any_init_locked): New inline function, does the locking.
850 Make use of these two throughout the code.
852 2005-08-15 Matthias Clasen <mclasen@redhat.com>
854 * glib/gbacktrace.c (g_on_error_stack_trace): Wait for
855 the child process and then simply return. This makes
856 The "S" option work as documented in g_on_error_query().
857 (#313125, Matthew F. Barnes)
859 * glib/gunicode.h: Update the link to Unicode category
860 values. (#313369, Behnam Esfahbod)
862 * glib/gqueue.c (g_queue_find_custom): Clarify docs
863 a little. (#311727, Tristan van Berkom)
865 * glib/abicheck.sh, gobject/abicheck.sh: Make the
866 check work on ia64 too, where some symbols ended up
867 in yet another section.